// Hunter Schafer, CSE 143
// This program counts the number of unique words in a large file.
// This program calculates the number of unique words in a file
// and prints out information about how long it took to run

import java.io.*;
import java.util.*;

public class UniqueWords {
    public static void main(String[] args) throws FileNotFoundException {
        System.out.println("Reading file...");
        Scanner input = new Scanner(new File("mobydick.txt"));

        long start = System.currentTimeMillis();

        int uniqueWords = countUnique(input);

        long end = System.currentTimeMillis();
        long elapsed = end - start;

        System.out.println("The file has " + uniqueWords + " words.");
        System.out.println("Took " + elapsed + " ms.");
    }

    // pre: input is not null
    // post: returns the number of unique tokens in input
    public static int countUnique(Scanner input) {
        // Hunter: Slow - List<String> words = new ArrayList<String>();
        // Hunter: Pretty fast and ordered - Set<String> words = new Treeset<String>();
        // Hunter: Fastest, but un-ordered
        Set<String> words = new HashSet<String>();

        // Hunter: I went ahead and simplified the logic in the method since
        // we are using a set (old code below)
        while (input.hasNext()) {
            String word = input.next();
            words.add(word);
        }

        return words.size();

        /* Hunter: We don't need to check if it contains with a Set since it ignores duplicates

        int count = 0;
        while (input.hasNext()) {
            String word = input.next();
            if (!words.contains(word)) {
                count++;
                words.add(word);
            }
        }
        return count;
        */
    }
}